New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 841413 link

Starred by 2 users

Issue metadata

Status: Duplicate
Merged: issue 835952
Owner:
Closed: May 2018
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

24% regression in blink_perf.canvas at 555727:555790

Project Member Reported by wangxianzhu@chromium.org, May 9 2018

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=841413

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=9383e5b8c03ce47b410c95450aea01627edfa52720360459f3fb10bcba9e9fbf


Bot(s) for this bug's original alert(s):

android-nexus5
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, May 11 2018

Cc: mtklein@chromium.org juliatut...@chromium.org gab@chromium.org dcreager@google.com eroman@chromium.org fmalita@chromium.org emaxx@chromium.org tsepez@chromium.org mmenke@chromium.org
Owner: mtklein@chromium.org
Status: Assigned (was: Untriaged)
📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14898877c40000

Reporting: Include include-subdomains in net log by dcreager@google.com
https://chromium.googlesource.com/chromium/src/+/210545ead75e144ff69728f0affeffb83a7ade5f

NetworkService:  Add HTTPS URL path stripping for PACs and quick check. by mmenke@chromium.org
https://chromium.googlesource.com/chromium/src/+/d476eddae62c071318339bb6b0d6c2fa2bd77f8e

use skcms' BUILD.gn by mtklein@chromium.org
https://chromium.googlesource.com/chromium/src/+/a3f77711c258e5661f14bd894e51177580ec623f

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Comment 4 by gab@chromium.org, May 11 2018

Cc: simonhatch@chromium.org sullivan@chromium.org
@sullivan/simonhatch : can pinpoint please stop CC'ing so many people on every regression? Landing mass refactorings and being a reviewer on many things means I'm CC'ed on these way too often (in fact in this instance I'm not even sure why I was CC'ed...).

As such I may start missing reports that are actually relevant to me.

We definitely shouldn't CC reviewers and sometimes when there are many identified culprits (i.e. low confidence) we probably shouldn't CC all owners either... Maybe add a tag and have a sheriff rotation to the assignment for regressions where the tool couldn't figure out a specific CL.

Productivity wise CC'ing everyone is like CC'ing no one IMO (and actually it's worse long-term because devs will get used to ignoring these).

Comment 5 by mmenke@chromium.org, May 11 2018

+1 to that, as someone who has learned to ignore these, and has even thought about a filter for them.
Cc: dtu@chromium.org
https://github.com/catapult-project/catapult/issues/4123

Dave, any chance we can bump the priority on this?
mmenke, gab: thanks for the feedback. Working on removing cc in  bug 842248 . After the issue with pinpoint not assigning to the author of the largest regression is fixed, we plan on having pinpoint post a comment on-CL when the CL causes a regression. Please comment on  bug 842248  if you'd like to see different behavior.
Cc: -mtklein@chromium.org
Mergedinto: 835952
Status: Duplicate (was: Assigned)

Sign in to add a comment